The bcnexus package & workflow

How the Python package is organised, and how one RunModel.run() call turns raw CSVs into a solved CLEWs model and results. Click any module or step for details.

Package structure

Modules at a glance

A simplified map of bcnexus/. The clews/ sub-package is the engine — a Runner (orchestrator) that drives a Builder plus helper modules; vis/ renders results.

bcnexus / (package root)

cli.py

utils.py

attributes_parser.py

constants.py

plots.py

bcnexus / clews / — the CLEWs engine

runner.pyRunner

builder.pyBuilder

model_structure.pyhelper

schema.pyhelper

sets_n_ratios.pyhelper

update_global_params.pyhelper

update_yearly_params.pyhelper

preprocess_data_Kotzur.pystorage

preprocess_data_Niet.pystorage

solver.pyhelper

datapackage.pyhelper

livestock.pyoptional

bcnexus / vis / — visualization

dashboard/Dash

plot_Energy · plot_Land · plot_Climate · …plots

vis_utils.py

Pick a module

Click any file on the left to see what it does and where it sits in the workflow.

How it connects to the workflow

One run() call, end to end

Creating a RunModel auto-creates a BuildModel (ClewsBuilder). Calling .run(build=True, threads=N) executes the whole pipeline. Click any step for detail.

① Configure

🎛️ Scenario config UI

utils.build_scenario_ui()

📦 model_builder_args

scenario · storage_algorithm · clustering

↓

② Instantiate clews/runner.py clews/builder.py

🧩 RunModel(**args)

creates ClewsBuilder inside

🚀 bcnexusRun.run(build=True, threads=N)

drives everything below

↓

③ Builder phase writes data/clews_data

🏗️ Build SETs & params + temporal clustering + storage case

model_structure · schema · sets_n_ratios · update_global/yearly_params · preprocess_data_Kotzur/Niet

↓

④ Runner phase — solve

📄 Prepare datafile

otoole · datapackage report

📝 Write LP

GLPK

🧮 Solve

solver.py · Gurobi / CBC

💾 Write solution

.sol · duals / shadow price

↓

⑤ Results & visuals

📊 Get result CSVs

get_result_csvs() · otoole

📈 Plots & dashboard

plots.main() · vis/dashboard

Data flow on disk

Where files live

Everything happens inside data/clews_data. The build creates clews_build_data/ with three sub-folders; results and plots land in their own trees. Set build=False to reuse a built structure and only re-cluster.

data/clews_data/clews_build_data/ ├─ input_csvs/ templates, updated per clews_builder.yaml ├─ inputs_csv_8760/ hourly (8760 h) profiles used for clustering └─ Model_<storage_algorithm>/<scenario>/ final datafiles results/clews/Model_<storage_algorithm>_<scenario>/ result CSVs (via otoole) vis/<scenario>/<timeslices>ts_<RUNTAG>/ saved figures